Think Graphical, Act Local: Distributed Inference for Robot Perception and Planning
Jana Pavlasek - Polytechnique Montréal
Jan. 9, 2026, 2:30 p.m. - Jan. 9, 2026, 3:30 p.m.
ENGMD 279
Hosted by: Hisu-Chin Lin
To achieve the long-promised general robot assistants capable of operating in our uncertain and unstructured world, we need efficient, reliable, and cooperative AI systems. My research explores advancements in distributed methods for efficient, adaptive, and robust inference for robotic perception and planning problems. Distributed inference leverages graphical models to break up complex problems into multiple smaller subproblems which can be solved in parallel. In this talk, I will present my work on distributed inference for challenging robotic problems including robot perception and multi-robot coordination. I will demonstrate how nonparametric representations of uncertainty help tackle high-dimensional problems with multiple diverse solutions efficiently. Finally, I will discuss how these ideas connect to robot foundation models and analyze their ability to generalize across tasks, environments, and embodiments.
Jana Pavlasek is an assistant professor in the Computer and Software Engineering Department (GIGL) at Polytechnique Montréal. She is also an Associate Academic Member at MILA, the Québec AI Institute. Her interests include robotic perception and planning under uncertainty and robot learning. Her research focus is on novel approaches which leverage probabilistic inference alongside machine learning for robust, autonomous robotic operation in unstructured environments. Jana earned her MSc and PhD in Robotics at the University of Michigan, and a Bachelor of Engineering from McGill University.
